The repetitive DNA structure of the human genome The recent completion of the human genome project gives today's scientists the privileged opportunity to provide, for the first and only time, a detailed comprehensive description of the structure of the human DNA sequence. Large parts of our genome remain relatively understudied, especially the repetitive DNA fractions, which account for greater than 45% of our total DNA sequence. Many important genomic turnover mechanisms contribute to the large accumulation of repetitive DNA in our genome, including genomic duplication, transposition, unequal crossing over, and gene conversion, which have a huge impact on the structure of our genome over the course of evolution. Therefore, we propose to undertake the first genome- wide survey and analysis of three distinct aspects of human repetitive DNA, by developing novel computer algorithms, genome analysis tools, and rigourous experimental approaches. 1) We propose to identify and characterize the complete catalogue of human inverted DNA repeats, which have been associated with many important genome functions such as DNA replication, meiotic crossover, and gene conversion. Our results have shown that the human X chromosome contains a preponderance of large highly homologous inverted repeats that contain testes genes. 2) We propose to investigate novel classes of tandemly repeated """"""""satellite"""""""" DNA that contain human transposons. These are organized in multiple large arrays primarily in the pericentromeric regions of chromosomes, where rapid chromosome evolution takes place. We have identified and characterized a large family of tandem repeats composed almost entirely of rearranged MaLR LTR transposons, found on 8 different human chromosomes in arrays as large as 70kb. 3) We propose to perform a genome-wide analysis of human transposable elements (TE's) by analyzing the large number of nested transposon clusters where newer TE's have transposed into older TE's. We have developed unique methodology and computer algorithms that can locate and index all such transposon clusters in the human genome, and can derive a relative chronological order of human TEs over the course of evolution. This represents a completely novel method of studying molecular evolution that is not dependent on the assumption of a constant mutation rate (molecular clock). The studies proposed in this application will facilitate both computational and biological approaches to genomics and provide a unique analysis of a large and relatively neglected portion of our DNA sequence.
